Oscillating crystal resonators can be used as very sensitive mass sensors in gas and liquid phase. It was shown by Sauerbrey (Sauerbrey, G., Z. Phys. 155 (1959), p. 206-222) that material deposited onto a resonator surface will change the resonators fundamental oscillation frequency proportional to the mass of the deposited material. Due to the extreme sensitivity of these resonators to changes in mass on their surfaces, oscillating crystal resonators can be employed to determine mass changes on a molecular level, and are often referred to as Quartz Crystal Microbalances or QCM. An oscillating crystal resonator generally consists of a thin plate of piezoelectric material, such as a quartz crystal wafer, with metal electrodes deposited on each face of the plate. Applying an electric field between the electrodes, or across the piezoelectric plate, causes a physical displacement in the piezoelectric material. Due to this “piezoelectric phenomenon” caused the by the electric current, steady oscillations of piezoelectric plates can be achieved through the application of a stable electric field in this manner. Once stable, changes in oscillation of the piezoelectric plates due to the addition or subtraction of mass from their surface can be quantified with great accuracy.
In 1980 researchers (Konash, P. L. and Bastiaans, G. J., Anal Chem. 52 (1980), p. 1929-1931), successfully utilized an oscillating crystal resonator as a sensor for measurement in the liquid phase. Over the years the use of oscillating crystal resonators as acoustic sensors in liquid phase applications have became very popular. Today there are thousands of literature publications documenting the application of oscillating quartz crystal resonators for use as liquid phase acoustic sensors. These sensors have been used to measure the presence and amount of chemical substances such as agricultural pesticides, toxins, and food additives in samples. A form of quartz sensor called a biosensor has been used to measure the presence and interactions of proteins, such as antibodies and hormones, nucleic acids, as well as pharmaceutical drugs, in liquid samples ranging from bodily fluids to organic solvents.
When operating an oscillating crystal resonators as a liquid phase sensor it is a requirement that the liquid sample interact with only one of the electrode coated surfaces on the resonator. The reasons for this are two-fold: 1) to eliminate electrical short-circuits between the electrodes of the resonator, and 2) to minimize loss of the resonator Q-factor (ratio of stored energy to dissipated energy in the piezoelectric plate) due to the liquids viscosity. To overcome the problem of creating a short-circuit between the electrodes the resonator mount design must isolate the back (driving/non-sensing) electrode from the front (sensing) electrode such that only the sensing electrode is exposed to the liquid under test. The viscosity of liquids can significantly dampen or even completely stop the oscillation of a crystal resonator. The higher the viscous load on the resonator the lower its sensitivity to changes in mass on the sensing surface. Thus, to minimize this dampening, it is preferable to expose only that part of the resonator required to perform the measurement to the liquid test sample, which again is the sensing electrode surface.
The earliest and most commonly used method for ensuring only one electrode of an oscillating crystal resonator came into contact with the liquid test sample was to sandwich, or “mount”, the resonator between a pair of rubber O-rings or gaskets. The O-ring or gasket on the sensing surface of the resonator was then interfaced with a well or cell such that the sample solution can be applied to that surface without being exposed to the other parts of the resonator. An example of this double O-ring resonator mounting configuration is disclosed in U.S. Pat. No. 5,135,852. As noted by the patent's author, a problem of this resonator mounting configuration is that the O-rings or gaskets exert fluctuating and non-reproducible pressure on the oscillating resonator, which directly impacts the sensitivity of the resonator. More specifically the author stated: in this sensor structure, the seals are placed at the edge of the sensor where the interference with its oscillations is minimal. However, this setup has the following drawbacks: 1) sensor response is strongly influenced by mounting pressure of the sample fluid within the cell, and this pressure adjustment is not readily reproducible; 2) during assembly, the quartz plate is handled directly resulting in the risk of damaging the fragile quartz plate; and 3) even when fixed firmly between the O-rings, distortions due to pressure fluctuations in the tested liquid, and expansion or contractions of the plate due to thermal changes will stress the sensor plate and cause friction between the sensor and the O-ring which in turn will result in decreased Q-factor and unsteady oscillations, i.e. noise sensor response.
This description highlights one of the basic problems in the use of oscillating crystal resonators in liquid-based applications. To be effectively employed as a liquid phase biosensor, the oscillating crystal resonator needs to be physically interfaced with liquid delivery unit so that only the sensing surface comes in contact with the liquid sample. However, any and all physical contact with the resonator, i.e., from the sample fluid, the mounting structure, e.g., the O-rings, etc., dampens its oscillation freedom and thus lowers it sensitivity and overall functionality. Similar to the example discussed above, the vast majority of resonator mounting designs to date have employed the use of elastic mounting seals or elastic adhesives to hold the resonator in place and create the liquid tight seal for the test sample chamber. The argument for the continued use of flexible mounting materials is the belief that the elasticity of the seal or adhesive will minimize its dampening of the resonator oscillation. However, while this elasticity in the mounting material may minimize dampening, it clearly has an impact on reproducibility between individual measurements on a single resonator and for measurements between different resonators.
To reduce the amount of stress on the mounted resonator, reduce the signal noise, and to improve reproducibility, various assemblies have subsequently been developed which use only one flexible O-ring or gasket and secure the resonator by pressing it against a solid mount, such as for example those sensor structures disclosed in PCT Application Nos. WO/2004/040268, WO/2002/061396, and WO/2002/012873. In all of these designs, the sensing (front) surface of the resonator is in physical contact with parts, solid or flexible, of the mounting assembly. However, because the resonator oscillations propagate out to the edge of the piezoelectric plate (even if the coated electrode does not reach to the edge of the plate), any component making physical contact with the sensing surface will impact the resonator response. Also, even slight distortions of the resonator from mechanical or thermal variations of the mounting assembly will result in noise added to the sensor response. As a result, none of the designs disclosed in the applications facilitates the construction of a reproducible sensor assembly.
More recently, another form of resonator mount designed to reduce sensor noise was disclosed in PCT Application No. WO/2002/047246. In contrast to the prior art sensor designs, in this design the resonator is placed with its non-sensing surface on a solid support surface and fixed to the surface with a flexible adhesive applied only along the edge of the resonator plate. Because the resonator is not exposed to direct and variable physical pressure from clamping, and the sensing electrode is not in contact with any mounting components, a significant noise reduction for liquid sensing applications is achieved, while at the same time the sensing electrode is isolated from the driving electrode, thereby preventing a short circuit across the resonator plate. However, this improved design still suffers from a number of problems that have been discussed in literature publications released following the patent filing. In particular, in practice it is very difficult to uniformly apply flexible adhesive around the edge of the very thin resonator plates (˜100 um) without depositing some amount of adhesive onto the sensing surface. It is also known that the resonator oscillations will reflect from the secured edge of the resonator plate, and thus the adhesive placed around the edge of the plate will still have an impact on the resonator response. Furthermore, while it is not physically pressed against the mounting substrate, the entire non-sensing surface of the resonator is in physical contact with the mounting substrate. Thus, even slight imperfections on the mounting surface will result in irregular stress on the resonator and noise in the sensor response.
Another problem with the use of oscillating crystal resonators in liquid based applications is the signal noise and response reproducibility issues arising from changes in hydraulic pressure on the resonator surface(s) during sample analysis. A primary requirement for the use of mounted oscillating crystal resonators in liquid based applications is the ability to expose sample-containing liquids to the sensing surface of the resonator as discrete volumes or plugs within a continuously flowing stream of a sample-less liquid. In a wide variety of applications, it is highly beneficial for the resonator sensing surface to be exposed to a liquid solution identical to that of the sample but without the sample before and after the sample solution is applied to the resonator sensing surface. It is also highly advantageous in a wide variety of applications that the exchanges from non-sample solution, to sample containing solution, and back to non-sample solution happen as instantaneously as possible. The processes involved with executing the additions and exchanges of sample and non-sample solutions to the sensing surface of the resonator result in pressure changes on the resonator surface(s), often resulting in non-reproducible stress on the resonators and dampening of the crystal oscillations. These stresses and dampening events can greatly diminish the sensitivity and reproducibility of the resonators as mass measurement sensors.
One approach to dealing with this problem has been to apply a counter-acting pressure on the backside of the resonator (driving electrode surface) to minimize irregular stress, deviations, and fluctuations on the sensing surface of the resonator. A problem with this type of solution comes from the design constraints of applying pressure on the backside of the mounted resonator due to the presence of the electrodes and the detection electronics connected to those electrodes. Also the intrinsic dampening of the resonator oscillations from exerting pressure on both faces of the resonator, and the difficulty in accurately matching the counter pressure on the driving electrode surface to the pressure being exerted on the sensing surface by the sample fluid or solution, make this practice highly impractical.
Thus, while the current resonator mounting designs are functional, they are far from optimal. Often the influence from erroneous signals in the form of response dampening, random noise, and drift created by the resonator mounting process lowers the sensitivity and reproducibility of the resonator to such a degree that the technology is not feasible for the majority of desired applications as a liquid based sensor. Clearly, to advance the use of this technology there is a requirement for an improved resonator mounting design. Ideally the mounting design would affix the oscillating crystal resonator to the mounting substrate in a highly stable, highly reproducible manner, fully isolating the liquid sample exposure to only the sensing surface of the resonator, while requiring only a minimal amount of physical contact between the mounting and sample delivery assembly and the oscillating resonator surfaces.